Oracle FAQ Your Portal to the Oracle Knowledge Grid
HOME | ASK QUESTION | ADD INFO | SEARCH | E-MAIL US
 

Home -> Community -> Usenet -> c.d.o.server -> Re: Database corrupted?

Re: Database corrupted?

From: Al Liu <angang.liu_at_saic.com>
Date: Fri, 11 Jan 2002 09:07:17 -0600
Message-ID: <3c3e8f2b@cpns1.saic.com>


The results of these two queries are
select count(*) from uet$;
COUNT(*)


      2987
select count (*) from fet$;
COUNT(*)


        53

"Stefan" <sroesch3_at_attbi.com> wrote in message news:Hrw%7.24596$762.223414_at_rwcrnsc54...
> Hi Al,
>
> you will never see ora-3113 in the alert.log as it is always a client
> problem. To check if you have problem with your extents run the following
> queries
>
> select count(*) from uet$;
> select count (*) from fet$;
>
> How many rows are returned?
>
> The other thing to check is set event 10046 with level 8 in one session
and
> check for the wait events.
>
> Regards
> Stefan
>
>
> Al Liu wrote:
>
> > ORA-03113 was not found in the alert.log, but error1092 was. About ten
> > minutes after I see the messages (they came up only a few seconds
apart),
> > I got my connection back. I'm not aware of NT shutdown in the past week.
I
> > suspect the problem have something to do with extents. Whenever I query
> > something about extents the query is very slow (usually takes a few
> > seconds, now take several minutes).
> >
> > "Ron Reidy" <rereidy_at_indra.com> wrote in message
> > news:3C3CC9DE.D7AA3C73_at_indra.com...
> >> Al Liu wrote:
> >> >
> >> > We have a ORACLE 8.0.5 database on NT4 served as data source for an
> >> > Intergraph GIS system. The size of database is ~2gb, the drive the
> > database
> >> > residing is 4 gb. In past three days, I have had the following
problems
> >> >
> >> > -some times when I try to connect database I got messages "ORA-03113:
> >> > end-of-file on communication channel" and " DBW0:terminating instance
> > due to
> >> > error 1092". Then after a while I can connect with either sqlplus or
> >> > svrmgr30. However, connection is much slower, taking 20-30 seconds.
> >> >
> >> > -queries generally runs slower than before, especially count(*)
query,
> > maybe
> >> > 10 times slower.
> >> >
> >> > -When I run snap8.sql usually take less than 2 minutes. It hangs when
I
> > run
> >> > it at retrieving datafile information. I checked the query, seems
like
> >> > dba_extents was the key of slowdown. I queried dba_extents alone and
> > noticed
> >> > that it stopped at four or five segments for a long time before
moving
> >> > forward again (total rows retrieved by the query is 2987). Query of
> >> > user_extents with a login of a user also run very slow.
> >> >
> >> > -The GIS system has real performance problem, some task used to take
> > less
> >> > than a minute may take more than a hour now.
> >> >
> >> > I checked V$datafile looks like all datafiles are online. I'm not
sure
> > all
> >> > above symptoms were related to the same problem. Please help me to
> > diagnose.
> >> >
> >> > Thanks.
> >> >
> >> > Al
> >> Did you check the alert log when you got ora-03113? How long has it
> >> been since your system (NT) was rebooted?
> >> --
> >> Ron Reidy
> >> Oracle DBA
>
Received on Fri Jan 11 2002 - 09:07:17 CST

Original text of this message

HOME | ASK QUESTION | ADD INFO | SEARCH | E-MAIL US